Skip to content

pelagornis/styleguide / .agents/skills/pelagornis-skill-authoring

pelagornis-skill-authoring

Verified

Create, review, or improve portable Agent Skills using the Pelagornis conventions and the open Agent Skills specification. Use when adding or maintaining a reusable agent workflow.

FileKindSize
SKILL.mdskill md1.3 KB
references/trigger-evals.mdmarkdown1.6 KB

references/trigger-evals.md

View on GitHub
# Trigger evaluation prompts

Use these prompts when checking automatic Skill selection. A matching prompt should activate the listed Skill; a non-matching prompt should not force that Skill. Ambiguous prompts need human review rather than a forced outcome.

| Skill | Should trigger | Should not trigger | Ambiguous |
|---|---|---|---|
| pelagornis-code-review | “Review this pull request for bugs and missing tests.” | “Implement the login form.” | “What changed in this branch?” |
| pelagornis-repository-audit | “Audit this repository and prioritize its security and maintenance risks. Do not modify files.” | “Fix the failing test.” | “How healthy is this project?” |
| pelagornis-git-workflow | “Prepare a Conventional Commit for the current changes.” | “Explain what Git rebase does.” | “Help me get this branch ready.” |
| pelagornis-test-debug | “Investigate this failing test and verify the smallest fix.” | “Review this diff for regressions.” | “The build is red; what should I inspect?” |
| pelagornis-docs-maintenance | “Update the documentation for the changed API and check links.” | “Create an API endpoint.” | “This feature needs a README update.” |
| pelagornis-skill-authoring | “Create a reusable Skill for our release checklist.” | “Add a permanent rule to AGENTS.md.” | “Should this process be a Skill?” |

Record the agent, version, prompt, whether the expected Skill was offered or selected, and any false positive or false negative. Revise descriptions when results are consistently too broad or too narrow.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexUnverifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall